The future

11:04 AM

I wasn't expecting so many to be against the change

Perhaps I didn't think about it very hard

3:48 PM

At this point I think SE could propose to send everyone $1000 no strings attached and people would still find something to complain about.

(not to denigrate the legitimate opinions of the people opposed to this particular change, I think there are some valid concerns)

Kevin B

4:05 PM

i mean

i don't want money from se

terdon - stop harming Monica

@Rubiksmoose heh, I certainly would object. First of all, because I wouldn't believe the no strings attached bit. I would just assume they are doing it for some legal reason that will become apparent down the line.

Kevin B

I see no reason for the change to happen, and also no real consequence of it happening. :shrug: i'm not happy about people asking awful questions getting more reward for doing so, but... meh, not like they're gonna use the privileges anyway.

terdon - stop harming Monica

@KevinB People asking awful questions will not be rewarded any more than they currently are. Upvotes are for good questions. If someone upvoted, then someone felt the question deserved the upvote.

Kevin B

Lol

people upvote awful querstions all the time, what you talkin bout

terdon - stop harming Monica

@KevinB They upvote questions you find awful.

I mean that if someone upvotes, then someone feels the question deserves an upvote.

You (or I) might not agree, but that's how voting works: it's a personal thing.

Kevin B

4:18 PM

Right, but clearly the weight is 1 sided

and this change only makes it worse

If anything... it will further demoralize the quality minded people who are left

terdon - stop harming Monica

I really don't see why. I think it far more likely to re-moralize people who ask good questions but feel their input isn't appreciated.

Kevin B

i mean, i don't see how you can't see the opposite is also true.

It's effectively reducing the value of one input vs another

terdon - stop harming Monica

No it isn't. That's precisely the point: increasing the value of one thing doesn't need to be read as decreasing the value of another. Their values are not defined with respect to one another, they are independent.

If I raise the price for beer, that doesn't mean I am lowering the price for water.

Kevin B

If increasing the value of your input makes you feel better, do you not see how reducing the value of your input may have the opposite effect?

terdon - stop harming Monica

@KevinB What is being reduced?

Kevin B

4:30 PM

beer and water are quite different. upvotes and downvotes are not

terdon - stop harming Monica

We're talking about questions and answers, not upvotes and downvotes.

Kevin B

the value of downvotes is. You can't cast both an upvote and a downvote, only one action can occur.

we're talking about both

You may only care about one side of it

terdon - stop harming Monica

But the change isn't up vs downvoting, it's answers vs questions.

Kevin B

It's increasing the power of upvotes on questions. which naturally decreases the power of downvotes (on questions). The end result is further weighted toward upvoting.

terdon - stop harming Monica

Power? Why? What we care about, as users, is that bad content goes down and good content goes up. This isn't actually linked to rep in any way. So this change will not affect how voting works: questions will still get downvotes just as quickly as now and still be removed from the front page just as quickly as now.

Kevin B

4:33 PM

Right

i'm not contesting that at all

terdon - stop harming Monica

So the "power" of the vote is the same. All that changes is that questions and answers are now the same in terms of the rep they gain their authors.

Kevin B

and since the beginning, one action has always been worth more rep than the other. One action has always gave people more "power" than the other took away. That's by design so that the target user doesn't feel like they can't gain any ground. This just pushes the ratio back to be equal to the ratio answers have been at.

At -2 +10, you can literally just spam answers all day, bad answers, and come out positively

they don't have to be good answers, or on good questions

you'll gain rep, because some small portion of them will be upvoted and not get deleted

terdon - stop harming Monica

Ah sorry, you meant up vs down, yes.

Kevin B

For answers, this is less of a problem

even poor answers are often helpful

terdon - stop harming Monica

But... you were talking about answers.

Did you mean "spam questions all day" above?

Kevin B

4:38 PM

no

i'm comparing the reward ratio between questions and answers

the actions people take, asking and answering

What's been normal is answering questions is the easiest and fastest way to earn rep. This change will not change that, because each question can have multiple answers. It's still going to be "easier" to go around answering as many questions as you can find because answers don't get closed. they only have to be moderately helpful to receive votes.

This change isn't going to take away the sting of being downvoted, it'l just give those who ask a lot rather than answer a lot more rep. more rep to apply bounties with, to get more upvotes, to apply more bounties,

terdon - stop harming Monica

Yes. All of which are good things, IMO. If anything, I find bad answers a much worse problem than bad questions since questions can be closed by relatively low rep users while answers can only be deleted and only by high rep users.

The main point though is that the only reason the scores were changed in the first place was to see if that change would lead to better questions. It didn't, it only leads to those people who do ask good questions being treated as second class citizens compared to those who answer. So undoing the change and bringing the two back into alignment sounds great.

Kevin B

it's inconsequential

terdon - stop harming Monica

After all, if introducing the change didn't bring about a rise in quality, why would undoing the change result in a drop? We have already demonstrated that there is no clear link between rep given and quality of questions.

Kevin B

why make such a change that will have such a massive influx of rep? because why not?

terdon - stop harming Monica

Especially since the vast majority of bad questions come from one-off accounts. The regulars, who are the only ones to even care about rep, learn to ask good ones.

@KevinB Because I honestly feel that questions are fully as important, if not more important, as answers and should be rewarded equally.

The sites would be useless without good questions to attract interesting answers.

Kevin B

4:46 PM

Questions don't need more motivation

terdon - stop harming Monica

So you've said. I disagree.

I also don't see rep as motivation, I see it as reward.

Kevin B

People ask questions because they need help. I don't want people who don't need help asking questions... and i don't expect this change to change that

so how would it in any way result in more questions or better questions?

it wont.

it's inconsequential

it's not going to encourage people who aren't asking to ask

terdon - stop harming Monica

@KevinB I don't think it will. It will result in the people who ask questions being adequately rewarded. I don't think it will affect question quantity or quality at all. That's precisely my point. You seem to think it's going to result in lower quality.

Kevin B

No, i don't think it's going to result in lower quality

Kevin B

5:39 PM

essentially... the only reason i see for this change is to undo a change that had no effect. due to recent events... i have no faith that it isn't being pushed for other motives that would clearly be negatively received if presented.

The change is inconsequential. It won't increase/decrease quality, it won't increase/decrease asking/answering rates, it won't increase/decrease voting or change the results of voting, it's... literally meaningless. So... why do it, if not for some other reason?

M.A.R.

8:21 PM

Wait, how do we know that?

I wasn't around when it was +10 for questions, doing . . . 13-year-old stuff and all that.

Kevin B

what

M.A.R.

@Kev @Ter how do you know decreasing the award hasn't had any important effects?

I mean, is there an objective way to measure it?

Tinkeringbell

10 year of SE questions? :P

Kevin B

I don't think we do, but that it had no effect seems to be the current "reason" for reversing it.

M.A.R.

OK, it just seems to me to be merely distinguishable from recency bias.

Or golden age fallacy

While my position is smartassness fallacy.

And your position, if you become snarky about my position, is anti-smartassness fallacy. And if you agree with it, syn-smartassness fallacy.

Zoe

8:25 PM

@M.A.R. Is there an objective way to measure the other way around?

M.A.R.

Fallacies all around.

Tinkeringbell

you're making my head hurt :(

M.A.R.

@Tinkeringbell yourlogicalfallacyis.com/the-fallacy-fallacy

Exercise makes you smarter. Brain exercise makes you smarterer

Tinkeringbell

Both make my head hurt though.

Zoe

Ther'es going to be people involved, and people ask different questions. Regardless of what vote value there is, there's going to be periods where bad questons are more likely (on a Q&A site, when there's pressure and school/work, but especially school)

M.A.R.

8:28 PM

Intuitively, I have some common sense arguments about whether it did or didn't change anything, but I don't feel like thinking so far enough as to reconcile them as to come up with any stance.

@Zoe Yeah, and there should be a way to eliminate or minimize the popularity 'effect' of attracting too much sand.

Kevin B

asking/answering/voting frequency always gets wonky around this time of year, with holidays, finals, winterbash, etc

M.A.R.

Sandstorms. The only real life way to look as awesome as The Batman animated series.

Zoe

@M.A.R. How? if the goal is to attract better questions, how can you effectively filter away legitimate attempts from "gimme teh rep"?

Kevin B

by giving more people the power to downvote

(which this change coincidentally will do)

M.A.R.

Ah, OK. I now have no idea if we're talking about the same thing.

Zoe

8:31 PM

yeah, but won't that scare away potential askers with good questions because of a fear of punishment? There's no proper way to manage it, and you've probably seen crappy questions get upvotes too.

Kevin B

oh, no, those questions weren't crappy, i'm just terrible at judging what is and isn't crappy

allegedly

M.A.R.

That one's worth analyzing too. Everyone has seen crappy questions getting upvoted, but are they statistical outliers or dangerously close to the norm?

Zoe

@KevinB GUILTY! ~~Bring in the guillotine~~

M.A.R.

Crappy meaning the bottom of the barrel, of course. Something like 99 percent of the people agreeing the question should be closed

Kevin B

think i'm somewhere around a 35% downvote to deletion ratio atm

Tinkeringbell

8:34 PM

@Zoe And an extra L ;)

M.A.R.

Guillotinel. It's an elf name

Zoe

My point is, you can't accurately measure if it has an effect or not because there's too many sources of error

M.A.R.

Sounds legit

Zoe

@Tinkeringbell Hi, I'm Zoe. I can't spell

xd

Tinkeringbell

No worries ;)

M.A.R.

8:35 PM

Ha, case in point

Edit back!

You just missed a big setup for a joke.

@Zoe No I'm talking about when it was reduced to +5.

There should be enough data unless that happened before 2010.

Kevin B

if i understood correctly the argument was that the data can't be trusted because it can be affected by too many outside factors

holidays, school cycles,

M.A.R.

Did it result in fewer questions? Did it result in better questions (distribution of votes)?

Zoe

@M.A.R. I'm talking about both

M.A.R.

@KevinB Wouldn't they even out given the time range is big enough?

Kevin B

possibly

M.A.R.

8:38 PM

If only I learned @rene's langu . . . I mean SQL and that stuff now I would have come up with six beautiful charts.

rene

u rang?

M.A.R.

I'm thinking, come up with a growth rate trend, compare large timescales before and after the reduction of question upvote rep, compare voting distributions, relative number of questions, and then we can objectively say whether it has had little effect back then so we could extrapolate whether the reverse will affect anything in the future

@rene No I'm just coming up with a thesis idea

Zoe

@M.A.R. eh...

M.A.R.

. . . I just bored myself.

Zoe

Recent closure trends have likely caused a decrease in questions, but there's always been a certain marketing push. If, some time after 2010, a major event hit that caused a significant increase in the site population (and that somehow lead to a question increase), punning that on a change in question vote value would be very, very wrong

M.A.R.

8:43 PM

It should be something Shog et al. come up with, if they delve into it as deeply as their close vote experiment analysis. I dunno, it's still a few years before I get good at that stuff.

@Zoe Then we'll choose the range carefully.

Anticlimactic times, like Summer of Love.

Someone go back in time and tell meta.SO folks just that.

Zoe

@M.A.R. Cherry picking the date range decreases accuracy

M.A.R.

Who said it should be 100 percent accuracy?

Zoe

There's still a difference between 90 and 40. Always opt for the highest accuracy

M.A.R.

Look at how many studies almost nailed their predictions with lower accuracy.

@Zoe I think if we do that we do get close enough to 90 percent accuracy, however that is measured.

It was just a rep change, dammit. It's not an immunodeficiency virus.

Spherical cow

A spherical cow is a humorous metaphor for highly simplified scientific models of complex real life phenomena. The implication is that theoretical physicists will often reduce a problem to the simplest form they can imagine in order to make calculations more feasible, even though such simplification may hinder the model's application to reality. The phrase comes from a joke that spoofs the simplifying assumptions that are sometimes used in theoretical physics. Milk production at a dairy farm was low, so the farmer wrote to the local university, asking for help from academia. A multidisciplinary...

Zoe

@M.A.R. it's not rocket science, it's just analyzing whether a change in irrelevant, digital points has a noticeable effect on a community.

Some people go nuclear at downvotes. You never know

/shrug

Kevin B

8:50 PM

i ran out of votes early today

Zoe

I haven't voted in a while

I did get 166 emails from SE though, and I'm expecting another 32 :]

M.A.R.

@Zoe Hey you're the one that's making this complicated! I'm just making it bor I mean scientific. sips leftover coffee

Zoe

@M.A.R. I always make things complicated. I overthink wayyyy too much

M.A.R.

I always overthink overthinking.

Kevin B

All along the rep change was just a test, and the test is concluding wed

M.A.R.

8:52 PM

We need to come up with a mathematical model that accurately defines borders for overthinking, thinking too much, thinking a little bit too much, and normal thinking

Kevin B

when did that change even happen? i think it was before i joined up

Zoe

@M.A.R. See, that's the beauty of overthinking - you don't need parameters, but you need meta overthinking

That way, you can have meta meta overthinking, and meta meta meta overthinking

M.A.R.

And a model for auto-economo-politico thinking, which is skillfully put to use by our beloved politicians. It doesn't involve thinking, but it has a rigorous theoretical framework

Kevin B

ah, 2010

i was definitely here

wait, no, that was a few months before i joined

is this still a thing?

> Item the second: after casting 300 votes, you cannot downvote non community wiki posts at more than a 2:1 ratio. Now, before you get up in arms, realize that this will affect very, very few users — on the order of about 6 users out of 100,000+ on Stack Overflow.

Zoe

nope

Kevin B

9:00 PM

i'm guessing not

because...

reasons

Zoe

I have 13k downvotes to 1.3k upvotes and I'm never prevented from downvoting more

Kevin B

i'm not quite at that ratio yet

give it a year or two

30k : 4k

M.A.R.

9:20 PM

There were people with very high down/up ratio, and witch hunts for such people etc. I guess that was created to prevent unnecessary drama and after things died down it wasn't all that important to keep around

Kevin B

As long as a reasonable percentage of your downvotes are resulting in deletion... fair game. but who gets to decide what reasonable is

M.A.R.

I say I am reasonable.

QED.

Kevin B

changing my ratios at this point is... not reasonable

would take years

assuming i didn't do so by ~~abusing~~ taking advantage of the system

M.A.R.

Abusing advantage of the taking.

Kevin B